One Python 2.1 idea

[Lieven Marchand]

"Tim Peters" <tim.one at home.com> writes:

> [Lieven Marchand]
> > I think Common Lisp shows that getting very good speed in Python is
> > quite feasable. The original CMUCL implementers weren't such a large
> > group. Their highly optimising compiler (which incidentally is also
> > called Python) has on occasion beaten FORTRAN at numerics. I don't
> > know why some people in the Python community think compiling Python is
> > such a problem.
> 
> The reasons vary, in large part depending on whether or not they've ever
> worked on a compiler <wink>.  Whatever, the fact is that there isn't a
> Python compiler, nor a credible plan for such a beast to come into
> existence.  If the CMUCL compiler group contained at least one person, it
> was infinitely larger than the group of people working on a Python compiler
> (Vyper notwithstanding).
> 

Perhaps because the community doesn't feel the need for a compiler
very strongly? I'd say the same effort that got you JPython could get
you a compiler.

> > Practically all the problems have been tackled and
> > solved 20 years ago in the Lisp community.
> 
> Excellent!  Then we would greatly appreciate it if you took one of their
> compilers, plugged in a Python parser, and released the result.  Sounds like
> it shouldn't take you more than a week or two, given that all the problems
> were solved so long ago.

I didn't say it wouldn't be a lot of work. I was reacting to some
people that feel that because Python is more dynamic than e.g. C it's
impossible to compile.

-- 
Lieven Marchand <mal at bewoner.dma.be>
Lambda calculus - Call us a mad club